A model-based controlled load-lock pumping strategy.
The load lock system controls pumping speed using a computer model and sensors to manage supersaturation ratio, addressing condensation issues and maintaining efficient throughput in semiconductor processing.
Patent Information
- Application Number
- JP2024523878
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-10-28
- Filing Date
- 2022-10-05
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-10-05
AI Technical Summary
Condensation during the pumping process in load locks leads to the formation of particles on workpieces, reducing yield and efficiency, and existing solutions that slow down the evacuation process to mitigate condensation negatively impact throughput.
A load lock system with a controller that manages the pumping speed to maintain the supersaturation ratio below a critical value using a computer model and sensors to adjust valve operation, ensuring efficient pump-down times without excessive condensation.
The system effectively minimizes condensation while maintaining throughput by controlling the supersaturation ratio, allowing pump-down times less than twice the theoretical minimum, thus reducing particle formation and improving yield.
Smart Images

Figure 2024543798000001_ABST
Abstract
Description
[Technical field]
[0001] This application claims priority to U.S. Patent Application No. 17 / 513,241, filed October 28, 2021, the disclosure of which is incorporated by reference in its entirety into this specification.
[0002] FIELD OF THE DISCLOSURE Embodiments of the present disclosure relate to load locks and, more particularly, to load locks that minimize condensation during the pumping process. [Background technology]
[0003] Loadlocks are used to transfer material from a first environment to a second environment when the pressures in the first and second environments are different. Loadlocks are often used as an interface between high vacuum and atmospheric environments. When transitioning from atmospheric environments, the loadlock opens to accept the material. When the loadlock closes, air is evacuated from the sealed chamber. The loadlock then opens to allow the material to be removed in the high vacuum environment. When transitioning from high vacuum environments, the loadlock introduces air to return the chamber to atmospheric pressure before opening.
[0004] In certain embodiments, as air is evacuated from the loadlock, the temperature within the loadlock may drop. If the temperature drops too much, condensation of water vapor may occur. This can occur if the supersaturation ratio exceeds a critical value at which the vapor condenses. Condensation can result in the formation of particles on the workpiece that may appear as defects after processing. This reduces yield and efficiency.
[0005] This problem can be mitigated by evacuating the loadlock at a slower rate. A slower change in pressure will result in a smaller drop in temperature, which may reduce the likelihood or amount of condensation. However, while effective at reducing particles, this approach implies that the workpieces are in the loadlock for longer periods of time, which impacts overall throughput.
[0006] It would therefore be beneficial to have a load lock that reduces the possibility of condensation but does not significantly increase the time of the pumping process. Furthermore, it would be advantageous if the load lock was adaptable to a variety of environmental and implant conditions. Summary of the Invention
[0007] A loadlock is disclosed in which the pumping speed is controlled to minimize the possibility of condensation. The loadlock is in communication with a vacuum pump and a valve. A controller is used to control the valve so that the supersaturation ratio in the loadlock does not exceed a predetermined threshold below a critical value at which the vapor will condense. In a particular embodiment, a computer model is used to generate a profile, which may be a pumping speed profile or a pressure profile, and the valve is controlled according to this profile. In another embodiment, the loadlock includes a temperature sensor and a pressure sensor. The controller can calculate the supersaturation ratio based on these parameters and control the valve accordingly.
[0008] According to an embodiment, a system for introducing a workpiece into a semiconductor processing tool is disclosed. The system includes a load lock, a vacuum pump, a valve between the load lock and the vacuum pump, and a controller including a memory element that stores a plurality of profiles, each associated with a corresponding relative humidity, where the controller is provided with an operating relative humidity, the controller selects one of the plurality of profiles, called a selected profile, based on the operating relative humidity, and controls the valve based on the selected profile. In a particular embodiment, the profile includes a pumping speed profile. In some embodiments, the profile includes a pressure profile. In a particular embodiment, the system includes a pressure sensor in communication with an interior of the load lock, and the controller controls the valve based on the pressure measured by the pressure sensor and the selected profile. In some embodiments, the profile is created using a computer model. In some embodiments, the computer model is created to maintain a supersaturation ratio in the load lock below a supersaturation ratio threshold that is below a critical value at which vapor condenses. In a particular embodiment, the computer model is created using differential equations. In some embodiments, the computer model is created using computational fluid dynamics.
[0009] According to another embodiment, a semiconductor processing system is disclosed comprising a semiconductor processing tool and the system described above.
[0010] According to another embodiment, a system for introducing a workpiece into a semiconductor processing tool is disclosed. The system includes a loadlock, a vacuum pump, a valve between the loadlock and the vacuum pump, and a controller with a memory element that stores a profile, the profile being created by a computer model to maintain the supersaturation ratio in the loadlock below a supersaturation ratio threshold below a critical value at which vapor condenses, and the pumpdown time is less than two times a theoretical minimum pumpdown time, the theoretical minimum pumpdown time being defined as the time to pump down the loadlock from atmospheric pressure to vacuum while the supersaturation ratio in the loadlock remains constant at a critical value, and the controller controls the valve based on the profile. In some embodiments, the profile includes a pressure profile. In certain embodiments, the system includes a pressure sensor in communication with the interior of the loadlock, and the controller controls the valve based on the pressure measured by the pressure sensor and the profile. In certain embodiments, the profile includes a pumping speed profile. In certain embodiments, the computer model is created using differential equations. In some embodiments, the computer model is created using computational fluid dynamics.
[0011] According to another embodiment, a system for introducing a workpiece into a semiconductor processing tool is disclosed. The system includes a loadlock, a vacuum pump, a pressure sensor in communication with an interior of the loadlock, a temperature sensor in communication with an interior of the loadlock, a valve between the loadlock and the vacuum pump, and a controller in communication with the pressure sensor and the temperature sensor, the controller calculating a supersaturation ratio in the loadlock based on information from the pressure sensor and the temperature sensor, and the controller controlling the valve based on the supersaturation ratio. In some embodiments, the controller controls the valve to maintain the supersaturation ratio below a supersaturation ratio threshold below a critical value at which steam condenses. In a particular embodiment, a humidity sensor, the supersaturation ratio is determined based on information from the humidity sensor. In a particular embodiment, the pumpdown time is less than twice the theoretical minimum pumpdown time, the theoretical minimum pumpdown time being defined as the time to pump down the loadlock from atmospheric pressure to vacuum while the supersaturation ratio in the loadlock remains constant at the critical value at which steam condenses.
[0012] According to another embodiment, a semiconductor processing system is disclosed comprising a semiconductor processing tool and the system described above.
[0013] For a better understanding of the present disclosure, reference is made to the accompanying drawings, which are incorporated herein by reference, in which: [Brief description of the drawings]
[0014] [Figure 1] FIG. 1 is a system schematic illustrating a load lock chamber according to one embodiment. [Figure 2A-2C] 1A-1C show pumping speed profiles, environmental parameters, and supersaturation ratios, each as a function of time, according to one embodiment. [Diagram 3] FIG. 13 illustrates a sequence that may be used to create a pumping speed profile, according to one embodiment. [Figure 4A-4C]6A-6C show pumping speed profiles, environmental parameters, and supersaturation ratios, each as a function of time, according to another embodiment. [Diagram 5] FIG. 11 illustrates a sequence that may be used to create a pumping speed profile according to a second embodiment. [Figure 6] FIG. 13 illustrates a sequence that may be used to create a pumping speed profile according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0015] As noted above, in certain embodiments, the workpieces are processed in a high vacuum environment. To process the workpieces, they are placed into a load lock, which is then pumped down to near vacuum conditions. The workpieces are then removed from the load lock and processed. The processed workpieces are then placed back into the load lock. The load lock is vented, and the workpieces are then removed at atmospheric pressure.
[0016] FIG. 1 illustrates a load lock 10 connected to a semiconductor processing tool 20. The load lock 10 allows semiconductor workpieces, such as wafers, to be loaded into the semiconductor processing tool 20 through a door 15 of the load lock 10. The semiconductor processing tool 20 may be any of a variety of processing equipment tools used in the semiconductor manufacturing industry. The semiconductor processing tool 20 may be a high vacuum tool in which the workpieces are processed at near vacuum or very low pressures. The semiconductor processing tool 20 may be an ion implanter, an etch tool, a deposition tool, or any of a variety of other tools used in the semiconductor device manufacturing industry. The type of semiconductor processing tool 20 is not limited by this disclosure.
[0017] Semiconductor workpieces, typically in the form of wafers, first pass through load lock 10 to reach the high vacuum processing section of semiconductor processing tool 20. Door 15 is first opened to allow a robot arm or other mechanical device to insert the unprocessed workpieces therein. Door 15 is then closed and the air within load lock 10 is removed by opening valve 30, which communicates with load lock 10 and vacuum pump 40. Once the chamber within load lock 10 is pumped to high vacuum, slot valve 19 opens to allow a robot arm to remove workpieces from load lock 10 and place them in the processing chamber of semiconductor processing tool 20 for processing. After the workpieces have been processed and returned to load lock 10, slot valve 19 is closed and the chamber of load lock 10 is vented to return the chamber to atmospheric pressure for removal from semiconductor processing tool 20.
[0018] In certain embodiments, various sensors may be used to measure parameters within the load lock. For example, there may be any combination of pressure sensors 16, humidity sensors 17, and / or temperature sensors 18. In certain embodiments, Pirani transducers may be used as the pressure sensors 16 and resistance temperature detectors (RTDs) may be used as the temperature sensors 18.
[0019] The controller 50 may be in communication with the sensors being used, which may include the pressure sensor 16, the humidity sensor 17, and the temperature sensor 18. Additionally, the controller 50 may be in communication with the valve 30. The valve 30 may be a variable control valve, in which case the valve can vary its opening to provide different pumping speeds. In other words, based on the size of the valve opening, the pumping speed achieved by the vacuum pump 40 varies. In certain embodiments, the valve 30 may be a butterfly valve or a ball valve.
[0020] The controller 50 includes a processing unit 51, such as a general purpose computer, a specially designed microcontroller, or any other suitable processing unit. The processing unit 51 is in communication with a memory element 52, also referred to as a computer memory, that stores instructions executed by the processing unit. These instructions enable the controller 50 to perform the functions described herein. During operation, the controller 50 can optionally receive pressure readings from the pressure sensor 16, humidity readings from the humidity sensor 17, and / or temperature readings from the temperature sensor 18, and output control signals to the valve 30 to adjust the pumping speed. The computer memory can also be used to store various parameter values during operation. Additionally, the memory element 52 can store one or more profiles, as described in more detail below.
[0021] As noted above, condensation may occur if the supersaturation ratio exceeds a critical value. Thus, in certain embodiments, the controller 50 creates a pumping speed profile that does not allow the supersaturation ratio to exceed its critical value. Specifically, the pumping speed profile is utilized when the loadlock transitions from a higher pressure condition, such as atmospheric pressure, to a lower pressure condition, such as less than 0.1 Torr.
[0022] According to one embodiment, the environmental conditions within the loadlock 10 may be approximated using a differential equation. Specifically, it is known that the air density within the loadlock may be expressed as: TIFF2024543798000002.tif14170In the above equation, ρ is the density of air in the load lock, Sp is the pumping speed, V is the volume of the load lock.
[0023] It is also known that pressure is directly related to density by the following formula: TIFF2024543798000003.tif13170In the above equation, P is the pressure in the load lock, R is the universal gas constant, T is the temperature inside the load lock, M is the molecular weight of the air in the load lock.
[0024] Furthermore, it is known that the temperature within the load lock can be expressed as: TIFF2024543798000004.tif14170In the above equation, T is the temperature inside the load lock, h is the convection heat transfer coefficient, A is the surface area of the inner surface of the load lock, C v is the specific heat of the air in the load lock at constant volume, T wall is the load lock wall temperature, γ is TIFF2024543798000005.tif14170, where Cp is the specific heat of air at constant pressure.
[0025] By solving these equations for density (ρ), pressure (P), and temperature (T), the supersaturation ratio can be found. The supersaturation ratio is defined as the partial pressure of water vapor divided by the saturation vapor pressure, where the saturation vapor pressure is a function of temperature. Curves showing the saturation vapor pressure as a function of temperature are readily available. Thus, the supersaturation ratio at any time is given by: TIFF2024543798000006.tif20170In the above formula, P v (t) is the vapor partial pressure as a function of time, P v (0) is the vapor partial pressure before pumping begins, which corresponds to the relative humidity, P(0) is the pressure before pumping begins, P(t) is the pressure in the load lock as a function of time. T(t) is the temperature in the load lock as a function of time. P sat (T(t)) is the saturation vapor pressure at the current temperature in the loadlock.
[0026] P vNote that (0) can be written as: P v (0)=RH(0)*P sat (0), In the above equation, RH(0) is the initial relative humidity, P sat (0) is the initial saturated vapor pressure.
[0027] Therefore, the supersaturation ratio can also be expressed as: TIFF2024543798000007.tif21170
[0028] P v Note that (0) is a function of the initial relative humidity in the chamber. After estimating the supersaturation ratio in the loadlock during pumping, the pumping speed can be altered to ensure that the supersaturation ratio does not exceed a critical value at which vapor will condense. In some embodiments, the supersaturation ratio does not exceed a supersaturation ratio threshold that is below the critical value.
[0029] One approach to ensure that the supersaturation ratio does not exceed the supersaturation ratio threshold is to utilize a closed loop controller, such as a PID controller, a PI controller, or a PD controller. In general, using a PID controller, the pumping rate may be defined as: TIFF2024543798000008.tif14170In the above formula, K P , K I , K D is the gain coefficient of the PID controller, e(t) is the error signal, e(t)=S target -S(t), where S target is the supersaturation ratio threshold.
[0030] In certain embodiments, the supersaturation ratio threshold may be a constant value. In other embodiments, the supersaturation ratio threshold may vary during the pump-down process. For example, the supersaturation ratio threshold may vary as a function of temperature or pressure within the load lock.
[0031] The coefficients of the PID controller may be selected so that the e(t) error function is non-negative and does not overshoot. In some embodiments, the controller may be simplified. For example, a PI or PD controller may be used by setting one of the coefficients to zero. Of course, other types of closed loop controllers may be employed.
[0032] In one embodiment, a computer model is developed that describes the environmental conditions in the loadlock as a function of pumping speed and time. The output from this computer model can then be used to calculate the supersaturation ratio in the loadlock 10. A pumping speed profile can then be created in a closed loop controller using the supersaturation ratio threshold and the calculated supersaturation ratio. For example, FIG. 2A shows a pumping speed profile generated using the computer model and equations shown above. FIG. 2B shows the environmental conditions in the loadlock 10 during pumpdown of the loadlock as calculated by the computer model. Specifically, the pressure 200 and temperature 210 in the loadlock 10 are presented. FIG. 2C shows the supersaturation ratio as a function of time. In this example, the closed loop controller was defined such that the supersaturation ratio does not exceed a predetermined supersaturation ratio threshold. This supersaturation ratio threshold can be any desired value, for example, 2.0 or less or 1.5 or less.
[0033] With reference to FIG. 2C, it can be seen that the pumping speed profile never exceeded the supersaturation ratio threshold. In this disclosure, an efficiency factor is defined. The efficiency factor relates two quantities. The first quantity is the area under a curve of supersaturation ratio versus time, such as the curve in FIG. 2C. The second quantity is the area of a rectangle with a height equal to the supersaturation ratio threshold and a width equal to the duration required for the loadlock to pump down from atmospheric pressure to 0.1 Torr. The efficiency ratio is defined as the ratio of the first quantity to the second quantity. In an ideal pumpdown profile, the environmental conditions in the loadlock would remain at the supersaturation ratio until the loadlock was completely pumped down. This would result in the first quantity being equal to the second quantity, i.e., the efficiency factor would be 1.0.
[0034] It should be noted that the efficiency factor is related to the theoretical minimum pump-down time. Specifically, the theoretical minimum pump-down time is achieved with a pumping speed profile with an efficiency factor of 1.0. In other words, the theoretical minimum pump-down time is defined as the time to pump the loadlock from atmospheric pressure to vacuum while the supersaturation ratio in the loadlock remains constant at a critical value. In some embodiments, the vacuum condition may be defined as less than 1 Torr. In other embodiments, the vacuum condition may be defined as less than 0.1 Torr.
[0035] A decrease in efficiency factor leads to an increase in actual pump-down time. In some embodiments, the systems and methods described herein create pump speed profiles that achieve an efficiency factor of more than 0.4, and the supersaturation ratio does not exceed a critical value. This may correspond to a pump-down time that is less than 2.5 times the theoretical minimum pump-down time. In certain embodiments, the systems and methods described herein create pump speed profiles that achieve an efficiency factor of more than 0.5, and the supersaturation ratio does not exceed a critical value. This may correspond to a pump-down time that is less than 2 times the theoretical minimum pump-down time. In other embodiments, the systems and methods described herein enable pump speed profiles that achieve an efficiency factor of more than 0.7, and the supersaturation ratio does not exceed a critical value. This may correspond to a pump-down time that is less than 1.5 times the theoretical minimum pump-down time.
[0036] It should be noted that if a PID controller is used, the pump speed profile may vary depending on the selection of KP, KI, and KD while still remaining below the supersaturation ratio threshold. Thus, while FIG. 2A illustrates one pump speed profile that may be used, there are many others. Furthermore, in this example, the pump speed profile was created using a PI controller. Other pump speed profiles may be generated using a PID or PD controller.
[0037] It should be noted that the pumping speed profile shown in Figure 2A can be reproduced by the controller 50 without the need for any feedback. Thus, in certain embodiments, the controller 50 operates in an open loop and does not use actual feedback from the load lock 10. Instead, the controller 50 simply matches the actual pumping speed at each point in time to the pumping speed profile shown in Figure 2A (or any other suitable pumping speed profile). This can be done by controlling the valve 30.
[0038] Thus, in some embodiments, the controller 50 does not utilize any input from the pressure sensor 16, the humidity sensor 17, or the temperature sensor 18. Instead, initial conditions such as the temperature and relative humidity of the walls of the load lock 10 may simply be assumed to be nominal values such as 22° C. and 50% RH, respectively.
[0039] Additionally, although the above describes the use of ordinary differential equations to simulate the environment within the loadlock 10, other modeling techniques may be used. For example, computational fluid dynamics (CFD) may be used to determine the environmental conditions, which may then be used in conjunction with a closed loop controller to calculate the pumping speed profile.
[0040] Thus, in this embodiment, the sequence of operations may be as shown in FIG. 3. First, a computer model is selected, as shown in box 300. This model may be an ODE (ordinary differential equations), CFD (computational fluid dynamics), or another model. This computer model is used to calculate the environmental conditions in the loadlock 10 as a function of pumping speed and time. Next, known parameters and assumptions may be input into the computer model, as shown in box 310. These known parameters and assumptions may include the surface area inside the loadlock, the temperature of the surfaces in the loadlock, the relative humidity, the volume in the loadlock 10, and various other parameters such as the natural convection heat transfer coefficient, the specific heat of air at constant volume, and the specific heat of air at constant pressure. It is noted that some computer models may not use all of these parameters. This computer model is used to calculate the temperature and pressure in the loadlock 10 at any time as a function of the pumping speed. From this information, the supersaturation ratio in the loadlock 10 at any time may also be calculated, as shown in box 320. A supersaturation ratio threshold is selected. Note that this may be done prior to the preceding processing. A pumping speed profile is then created using the above equations, the supersaturation ratio and the supersaturation ratio threshold, and a closed loop controller, such as a PID, PI, or PD controller, as shown in box 330. Specifically, for each time interval, the closed loop controller calculates a desired pumping speed based on the calculated supersaturation ratio and the supersaturation ratio threshold. This new value is then input into the differential equation to calculate a new temperature and pressure. This new temperature and pressure are then used to calculate a new supersaturation ratio.
[0041] It should be noted that boxes 300-330 may be performed by controller 50 or by a different processing unit. If performed by controller 50, the closed-loop controller may be implemented in software within controller 50 or may be a separate hardware block within controller 50. If boxes 300-330 are performed by a separate processing unit, the processing unit may be a dedicated PID controller or may be a computer programmed to perform the operations of the closed-loop controller. In either embodiment, the resulting pumping speed profile is stored in controller 50, as shown in box 340.
[0042] It should be noted that although the use of a closed loop controller is discussed above, other techniques may also be used. For example, the pumping speed profile may be calculated manually based on a computer model.
[0043] Finally, as shown in box 350, the controller 50 adjusts the valve 30 so that the actual pumping speed matches the pumping speed from the pumping speed profile. As noted above, this can be considered open loop control because the actual pressure and temperature of the load lock 10 are not used by the controller 50 when it adjusts the valve 30.
[0044] As a variation of FIG. 3, multiple different pumping speed profiles may be generated using the process shown in box 330, with each pumping speed profile based on a different relative humidity. For example, as discussed above, the formula for the supersaturation ratio depends on the initial relative humidity in the load lock 10. Different values of relative humidity will therefore produce different results. Thus, in another embodiment, multiple pumping speed profiles, each designed for a given relative humidity, are stored in the controller 50 in box 340. The controller 50 is then provided with the relative humidity in the load lock 10. This may be done using the humidity sensor 17 or may be input by an operator. The controller 50 may then utilize the pumping speed profile generated with the relative humidity closest to the input relative humidity. In a particular embodiment, the pumping speed profile selected may be the one generated with the lowest relative humidity that is equal to or greater than the input relative humidity. For example, if the controller 50 has stored five pumping speed profiles established at relative humidity of 20%, 30%, 40%, 50% and 60%, and the input humidity is 37%, the controller 50 may select the pumping speed profile generated at 40%. In this way, the supersaturation ratio in the load lock 10 will be less than that envisaged in the selected profile.
[0045] Using the relative humidity as an input can help improve throughput. In Figures 2A-2C, a relative humidity of 50% is assumed. This resulted in a pump-down time of about 65 seconds. In Figures 4A-4C, a relative humidity of 20% is assumed. Figure 4A shows the pumping speed profile as a function of time. Figure 4B shows the environmental parameters including pressure 220 and temperature 230 as a function of time. Figure 4C shows the supersaturation ratio as a function of time. Note that the pump-down time was reduced by about 25 seconds because the starting relative humidity was much lower. This can result in an increase in throughput through the load lock of over 30%.
[0046] Thus, in this embodiment, the controller 50 stores multiple pumping speed profiles and selects one of these profiles based on the input parameters. Although the above example assumes that relative humidity is the input parameter, other parameters can be input, including the temperature of the load lock 10.
[0047] It is further noted that a temperature profile or pressure profile may also be generated based on the pumping speed profile. For example, rather than directly controlling the pumping speed, the controller 50 may control the valve 30 to achieve a desired pressure profile. Thus, in addition or alternatively, multiple pressure profiles may be stored in the controller 50 and used to control the valve 30.
[0048] In another embodiment, the controller 50 utilizes feedback from the pressure sensor 16, humidity sensor 17, and / or temperature sensor 18 to adjust the valve 30. For example, if the pressure sensor 16, humidity sensor 17, and temperature sensor 18 are present, the controller 50 can easily calculate the supersaturation ratio at any time. This supersaturation ratio can then be utilized by a closed loop controller, such as a PID, PI, or PD controller, to determine the desired pumping speed at that time.
[0049] 5, the controller 50 receives as inputs the humidity, temperature, and pressure, as shown in box 500. The controller 50 then calculates the current supersaturation ratio, as shown in box 510. The calculated supersaturation ratio and the supersaturation ratio threshold are then used as inputs to a closed loop controller, such as a PID, PD, or PI controller, to calculate a desired pumping rate, as shown in box 520. The controller 50 then adjusts the valve 30 to achieve this pumping rate, as shown in box 530. The sequence is then repeated until the pressure reaches the desired level, at which point the throttle valve 19 may be opened.
[0050] Thus, in this embodiment, the controller 50 adjusts the pumping speed using closed loop control. Specifically, the controller 50 measures the environmental conditions within the load lock 10, calculates the supersaturation ratio based on those conditions, and then adjusts the valve 30 accordingly.
[0051] In yet another embodiment, the controller 50 may utilize a hybrid approach that uses a computer model in conjunction with one or more sensor values to determine the environmental conditions within the load lock 10. For example, as seen in FIG. 2B and FIG. 4B, pressure changes relatively slowly compared to temperature. Therefore, it may be possible to utilize the pressure sensor 16 to monitor the pressure. The measured pressure value may then be used as an input to the computer model. For example, the measured pressure and the current pumping speed may be used to determine the current temperature within the load lock 10. The measured pressure and the calculated temperature may be used to calculate a supersaturation ratio. The controller may then use this supersaturation ratio to calculate a new pumping speed.
[0052] 6 illustrates the sequence associated with this approach. One or more environmental conditions within the load lock 10 are measured and provided to the controller 50, as shown in box 600. Based on the measured conditions and a computer model, the remaining environmental conditions may also be calculated, as shown in box 610. Once all environmental conditions have been measured or calculated, the supersaturation ratio may be computed, as shown in box 620. The calculated supersaturation ratio and the supersaturation ratio threshold are then used as inputs to a closed loop controller, such as a PID, PD, or PI controller, to calculate a desired pumping rate, as shown in box 630. The controller 50 then adjusts the valve 30 to achieve this pumping rate, as shown in box 640. The sequence is then repeated until the pressure reaches the desired level, at which point the slot valve 19 may be opened.
[0053] Thus, in this scenario, the controller 50 uses both the measured and calculated environmental conditions to calculate the supersaturation ratio. The calculated supersaturation ratio and the supersaturation ratio threshold are then used to determine an appropriate pumping speed. The controller 50 then adjusts the valve 30 depending on the pumping speed.
[0054] In yet another embodiment utilizing a hybrid model, the pressure sensor 16 may be employed. In this embodiment, a computer model is used to create a pumping speed profile, as described above and shown in FIG. 3. Based on this pumping speed profile, a pressure profile may then be created, such as the graph of pressure 200 shown in FIG. 2B. This pressure profile is then stored in the controller 50. The controller 50 then compares the pressure measured by the pressure sensor 16 to the pressure profile. Based on this difference, a new pumping speed may be determined. This pumping speed may be generated using a closed loop controller as described above, except that an error function may be defined as the difference between the pressure in the pressure profile and the actual pressure in the load lock measured by the pressure sensor 16. Thus, in this embodiment, the controller utilizes a theoretical pressure profile in conjunction with the actual pressure measurements to control the valve 30.
[0055] In a variation of this approach, multiple pressure profiles, each designed for a given relative humidity, are stored in the controller 50. The controller 50 is then fed with the relative humidity in the load lock 10. This may be done using the humidity sensor 17 or may be input by an operator. The controller 50 can then utilize the pressure profile that was generated with the relative humidity closest to the input relative humidity.
[0056] The above-described embodiments of this application may have many advantages. In a high relative humidity environment, the supersaturation ratio in the loadlock may rise to an unacceptable level, resulting in water particles adhering to the workpiece. These water particles may cause defects during processing of the workpiece. However, simply slowing down the pumping speed to maintain a lower supersaturation ratio in the loadlock may cost valuable time and impact throughput. Using the supersaturation ratio in the loadlock (either measured or calculated using a computer model), the pumping speed profile may be adjusted to ensure that water particles do not form and that the duration used to pump down the loadlock remains reasonable.
[0057] For example, in certain current systems, a pumping speed profile is used. When the relative humidity is high, the pumping speed profile is modified to prevent condensation by reducing the pumping speed. In one test, the pumping down time was extended by 100 seconds to eliminate condensation. By utilizing a pumping speed profile based on maintaining the supersaturation ratio below a critical value, faster pumping down times can be achieved. For example, the faster pumping down times can eliminate the additional 100 seconds mentioned above. Furthermore, as mentioned above, the systems and methods described herein can enable the load lock to be pumped down in less than twice the theoretical minimum pumping down time while preventing the supersaturation ratio in the load lock from exceeding the supersaturation ratio threshold. In some embodiments, the pumping down time is less than 1.5 times the theoretical minimum pumping down time. In contrast, current systems typically have pumping down times that are more than three times the theoretical minimum pumping down time.
[0058] The present disclosure is not limited in scope by the specific embodiments described herein. Indeed, in addition to those described herein, various other embodiments of the present disclosure and modifications thereto will be apparent to those skilled in the art from the foregoing description and the accompanying drawings. Thus, such other embodiments and modifications are intended to fall within the scope of the present disclosure. Furthermore, although the present disclosure has been described herein in the context of specific implementations in specific environments for specific purposes, those skilled in the art will recognize that its usefulness is not limited thereto, and that the present disclosure may be beneficially implemented in any number of environments for any number of purposes. Thus, the claims set forth below should be construed in light of the full scope and spirit of the present disclosure as described herein.
Claims
1. A load lock and A vacuum pump; a valve between the load lock and the vacuum pump; a controller including a memory element that stores a plurality of profiles, each associated with a corresponding relative humidity, the controller being provided with an operating relative humidity, the controller selecting one of the plurality of profiles, referred to as a selected profile, based on the operating relative humidity, and controlling the valve based on the selected profile; A system for introducing a workpiece into a semiconductor processing tool, comprising:
2. The system of claim 1 , wherein the profile comprises a pumping speed profile.
3. The system of claim 1 , wherein the profile comprises a pressure profile.
4. 4. The system of claim 3, further comprising a pressure sensor in communication with an interior of the load lock, the controller controlling the valve based on a pressure measured by the pressure sensor and the selected profile.
5. The system of claim 1 , wherein the profile is created using a computer model.
6. The system of claim 5 , wherein the computer model is developed to maintain a supersaturation ratio in the load lock below a supersaturation ratio threshold below a critical value at which vapor condenses.
7. The system of claim 5 , wherein the computer model is developed using differential equations.
8. The system of claim 5 , wherein the computer model is created using computational fluid dynamics.
9. A semiconductor processing tool; The system according to claim 1 . A semiconductor processing system comprising:
10. A load lock and A vacuum pump; a valve between the load lock and the vacuum pump; a controller including a memory element for storing a profile, the profile being generated by a computer model to maintain a supersaturation ratio in the loadlock below a supersaturation ratio threshold below a critical value at which vapor condenses and to have a pumpdown time less than twice a theoretical minimum pumpdown time, the theoretical minimum pumpdown time being defined as the time to pump down the loadlock from atmospheric pressure to vacuum while the supersaturation ratio in the loadlock remains constant at the critical value, and the controller controlling the valve based on the profile; A system for introducing a workpiece into a semiconductor processing tool, comprising:
11. The system of claim 10 , wherein the profile comprises a pressure profile.
12. 12. The system of claim 11, further comprising a pressure sensor in communication with an interior of the load lock, the controller controlling the valve based on a pressure measured by the pressure sensor and the profile.
13. The system of claim 10 , wherein the profile comprises a pumping speed profile.
14. The system of claim 10 , wherein the computer model is developed using differential equations.
15. The system of claim 10 , wherein the computer model is created using computational fluid dynamics.
16. A load lock and A vacuum pump; a pressure sensor in communication with an interior of the load lock; a temperature sensor in communication with the interior of the load lock; a valve between the load lock and the vacuum pump; a controller in communication with the pressure sensor and the temperature sensor, the controller calculating a supersaturation ratio in the load lock based on information from the pressure sensor and the temperature sensor, and the controller controlling the valve based on the supersaturation ratio; A system for introducing a workpiece into a semiconductor processing tool, comprising:
17. 17. The system of claim 16, wherein the controller controls the valve such that the supersaturation ratio remains below a supersaturation ratio threshold below a critical value at which vapor condenses.
18. 20. The system of claim 17, further comprising a humidity sensor, and wherein the supersaturation ratio is determined based on information from the humidity sensor.
19. 17. The system of claim 16, wherein the pump-down time is less than twice a theoretical minimum pump-down time, said theoretical minimum pump-down time being defined as the time to pump down said load lock from atmospheric conditions to vacuum conditions while said supersaturation ratio in said load lock remains constant at a critical value at which vapor condenses.
20. A semiconductor processing tool; A system according to claim 16. A semiconductor processing system comprising:
Citation Information
Patent Citations
Exposing equipment and method for manufacturing device
JP2005191494A
Vacuum processing apparatus
JP2011228397A
